Neutron diffraction structure study of Er and Yb doped YAl
3(BO
3)
4E. Sváb
⇑, E. Beregi, M. Fábián, Gy. Mészáros
Research Institute for Solid State Physics an Optics, P.O.B. 49, H-1525 Budapest, Hungary
a r t i c l e i n f o
Article history:
Received 5 September 2011
Received in revised form 18 January 2012 Accepted 2 March 2012
Available online 31 March 2012
Keywords:
Rare earth aluminium borates Non-linear optical material Neutron diffraction Rietveld refinement Atomic distances
a b s t r a c t
Neutron diffraction structure study has been performed on YAl3(BO3)4(YAB), on doped Y0.88Er0.12Al3
(BO3)4, Y0.5Er0.5Al3(BO3)4, Y0.5Yb0.5Al3(BO3)4and on co-doped Y0.84Er0.01Yb0.15Al3(BO3)4compositions. It was established that the doped compounds are isostructural to YAB. The neutron diffraction pattern have been be fitted in space groupR32 using the triple hexagonal Wyckoff notation. Both Er3+and Yb3+ions occupy the Y3+(3a) sites and not the Al3+(9d) sites, as it was suggested previously. The lattice parameters are decreasing with increasing amount of the dopant elements. Slight changes are revealed in the posi- tional parameters and interatomic distances with increasing concentration of the dopant ions. For the co-doped Y0.84Er0.01Yb0.15Al3(BO3)4the changes are more significant than for the doped YAB compounds with only one type of dopant element, Er or Yb.
Ó2012 Elsevier B.V. All rights reserved.
1. Introduction
Yttrium aluminium borate YAl3(BO3)4(YAB) is well known as non-linear optical material and has important applications in laser engineering[1]. According to its absorption, luminescence, optical and non-linear optical properties, YAB is a suitable host for laser-active dopants and has suitable sites for different ions e.g.
rare-earth ions (Er3+, Nd3+, Yb3+ and La3+) or other doping ions e.g. transition elements (Cr3+, Fe3+ and Ga3+) at Al3+ site. Some authors, however, reported[2]that Er3+prefers the Al3+site due to its relatively small radius.
Efficient laser emission, self-frequency doubling and self- frequency summing have already been demonstrated for Nd and Yb dopants in YAB[3,4].
Recently, efficient laser action was demonstrated with Yb(11%), Er(1.5%)-activated YAB borate. It was reported to act as a very promising laser crystal for the spectral range between 1.5 and 1.6
l
m[5]. Good laser efficiency of YAB requires high concentra- tion of the active dopants. However, sometimes with increase of dopant concentration, the spectroscopic line broadening can be ob- served accompanied with weak satellites around the main lines which can be attributed to weak Er–Er interaction, as it was dis- cussed in Ref.[6]. The knowledge of the details of crystal structure including the crystallographic parameters is essential for the calcu- lation of crystal field parameters and for characterisation of the optical properties of crystals. In a previous study we have reported the crystallographic results on Ga-doped YAB structure[7].The crystal system of YAl3(BO3)4 is trigonal with space group s.g.R32 (No. 155)[7,8]. For the description of the compounds un- der investigation the triple hexagonal unit cell has been applied, the number of formula units in the unit cell is Z= 3. The Y, Al and B atoms are coordinated by oxygen ions in the respective tri- gonal-prismatic, octahedral and planar- triangle geometry. Follow- ing the Wyckoff notation, the unit cell contains 12 B atoms in two different sites B(1) in (3b) and B(2) in (9e) positions, 9 Al atoms in three differently oriented but energetically equivalent (9d) posi- tions, 3 Y atoms in (3a) positions and 36 O atoms in three different sites O(1) in (9e), O(2) in (9e) and O(3) in (18f) positions.
The aim of this neutron diffraction study is to investigate the ef- fect of Er or/and Yb substitution on the crystallographic parame- ters of the studied crystals isomorphous to YAB, like lattice parameters, positional parameters and to calculate the interatomic distances, which maybe important for the analyses of optical characteristics.
2. Experimental 2.1. Samples
For this study we have prepared single crystals with composi- tions YAl3(BO3)4, doped Y0.88Er0.12Al3(BO3)4, Y0.5Er0.5Al3(BO3)4, Y0.5Yb0.5Al3(BO3)4and co-doped Y0.84Er0.01Yb0.15Al3(BO3)4. Because of their incongruent melting, the crystals were grown by top- seeded high temperature solution method (HTSSG) from K2Mo3O10-B2O3flux in 1150–1000°C temperature range[9]. The incorporation of Er and Yb into Y-lattice sites is relatively easy due to the similarity of the ionic radii, being 0.90, 0.89 and 0925-3467/$ - see front matterÓ2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.optmat.2012.03.003
⇑ Corresponding author. Tel.: +36 1 3922634.
E-mail addresses: svab@szfki.hu (E. Sváb), beregi@szfki.hu (E. Beregi), fabian@szfki.hu(M. Fábián),mgy@szfki.hu(Gy. Mészáros).
Optical Materials 34 (2012) 1473–1476
Contents lists available atSciVerse ScienceDirect
Optical Materials
j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / o p t m a t
0.87 Å for Y, Er and Yb, respectively (seeTable 2in Ref.[10]). The Er segregation coefficient was around unity and the compositions were close to stoichiometry[11].
The single crystals were powdered into fine grains to obtain good powder pattern for the neutron diffraction measurements.
2.2. Neutron diffraction experiments
The neutron diffraction measurements have been performed at the 10 MW Budapest research reactor using the ‘PSD’ neutron dif- fractometer [12] with monochromatic wavelength k= 1.0577 Å.
Fig. 1.Neutron diffraction pattern (k= 1.0577 Å) and Rietveld refinement in space groupR32 of doped YAB compounds: (a) YAl3(BO3)4(RB= 4.7%); (b) Y0.88Er0.12Al3(BO3)4
(RB= 3.7%); (c) Y0.5Er0.5Al3(BO3)4(RB= 5.9%); (d) Y0.5Yb0.5Al3(BO3)4(RB= 4.4%); (e) Y0.84Er0.01Yb0.15Al3(BO3)4(RB= 11.7%).
1474 E. Sváb et al. / Optical Materials 34 (2012) 1473–1476
The absorption of natural boron for neutrons is extremely high [13], therefore a cylindrical vanadium can of relatively small diam- eter, 5 mm, was used, which gave a transmission value of about 50%. It should be noted, that the high neutron absorption of natural boron might be avoided in several cases by preparing the samples from11B-isotope. The authors have applied the isotope substitu- tion technique for preparing borosilicate glasses for neutron dif- fraction study[14 and references therein]. In this study we have considered the possibility to prepare the YAB crystals from11B-iso- tope. The problem, however, in this case is that, although the crys- tal itself is small, the entire amount of the flux has to be prepared from11B-isotope, which would be very expensive and a great loss proceeds of the highly valuable isotope. Therefore we decided to work with samples containing natural boron.
Neutron diffraction pattern measured in the 2h= 10–90°range were refined by Rietveld method using the program suite Fullprof [15].
3. Results and discussion
The overall run of the experimental neutron diffraction data of the investigated compounds is very similar to each other, even by large amount 50at% of dopant element, as it is illustrated inFig. 1.
This observation indicates that Er or Yb doped YAB compounds are isotructural to YAl3(BO3)4, the powder pattern can be fitted in hex- agonal s.g.R32. For the Rietveld refinement calculations we started from the YAB parameters given in hexagonal s.g.R32, as revealed in our previous work[7]. The fit proved to be fairly good for each compound, which is characterised by theRBragg(RB) factor, as indi- cated inFig. 1. For the co-doped compound theRBfactor is larger, which may be the reason of the small amount of the specimen leading to worse statistics of the experimental data.
The unit cell parameters show a slight but significant decrease with increasing Er, Yb concentration, as they are collected inTable 1. This variation is consistent with the slight decrease of atomic ra- dii. It is noteworthy, that the decrease of the lattice parameters for the co-doped Y0.84Er0.01Yb0.15Al3(BO3)4is greater than for the YAB compounds doped with only one type of element (Er or Yb). The unit cell volumes are also given inTable 1, which show a decrease with increasing concentration of the dopant element in accordance with the decrease of lattice parameters.
As the first step of data evaluation procedure, we have per- formed refinements to answer the question whether Er or Yb sub- stitute the Y (model-1) or Al (model-2) sites. This is important, because in Ref.[2]it has been reported, that Er prefers the Al sites.
Neutron diffraction is a suitable method to answer this question, Table 1
Lattice parameters (Å) and unit cell volume (Å3) in hexagonal space groupR32.
Composition a c Vunit cell
YAl3(BO3)4 9.293(2) 7.236(8) 541.26
Y0.88Er0.12Al3(BO3)4 9.278(6) 7.224(6) 538.66 Y0.50Er0.50Al3(BO3)4 9.267(8) 7.216(9) 536.82 Y0.50Yb0.50Al3(BO3)4 9.267(6) 7.211(5) 536.40 Y0.84Er0.01Yb0.15Al3(BO3)4 9.263(2) 7.211(9) 535.92
Table 2
Refined positional parameters in space groupR32 for YAB and Er doped compounds.
Atom Wyckoff position YAl3(BO3)4 Y0.88Er0.12Al3(BO3)4 Y0.50Er0.50Al3(BO3)4
x/a y/b z/c x/a y/b z/c x/a y/b z/c
Y(Er) (3a) 0 0 0 0 0 0 0 0 0
Al (9d) 0.554(4) 0 0 0.554(2) 0 0 0.555(5) 0 0
B(1) (3b) 0 0 0.5 0 0 0.5 0 0 0.5
B(2) (9e) 0.454(5) 0 0.5 0.442(2) 0 0.5 0.442(1) 0 0.5
O(1) (9e) 0.850(8) 0 0.5 0.850(8) 0 0.5 0.852(7) 0 0.5
O(2) (9e) 0.589(7) 0 0.5 0.590(6) 0 0.5 0.592(9) 0 0.5
O(3) (18f) 0.449(3) 0.148(9) 0.515(6) 0.449(8) 0.149(6) 0.521(2) 0.450(2) 0.150(7) 0.522(8)
Table 3
Refined positional parameters in space groupR32 for YAB compounds doped with Yb and co-doped with Er and Yb.
Atom Wyckoff position Y0.50Yb0.50Al3(BO3)4 Y0.84Er0.01Yb0.15Al3(BO3)4
x/a y/b z/c x/a y/b z/c
Y(Er,Yb) (3a) 0 0 0 0 0 0
Al (9d) 0.558(5) 0 0 0.554(3) 0 0
B(1) (3b) 0 0 0.5 0 0 0.5
B(2) (9e) 0.441(4) 0 0.5 0.442(1) 0 0.5
O(1) (9e) 0.853(1) 0 0.5 0.855(6) 0 0.5
O(2) (9e) 0.595(9) 0 0.5 0.589(3) 0 0.5
O(3) (18f) 0.448(8) 0.147(8) 0.522(9) 0.443(4) 0.153(4) 0.519(1)
Table 4
Interatomic distances [Å].
Neighbouring atom pairs YAl3(BO3)4 Y0.88Er0.12Al3(BO3)4 Y0.50Er0.50Al3(BO3)4 Y0.50Yb0.50Al3(BO3)4 Y0.84Er0.01Yb0.15Al3(BO3)4
Y(Er,Yb)–Y(Er,Yb) 6 5.873(5) 5.873(2) 5.866(7) 5.865(8) 5.863(6)
Y–O(3); 6 2.318(2) 2.317(3) 2.316(0) 2.330(2) 2.324(5)
Al–O(1); 2 1.916(1) 1.913(6) 1.924(7) 1.924(2) 1.941(5)
Al–O(2); 2 1.928(3) 1.942(4) 1.917(6) 1.880(3) 1.946(2)
Al–O(3); 2 1.859(3) 1.849(7) 1.846(2) 1.871(0) 1.789(4)
B(1)–O(1); 3 1.383(4) 1.389(7) 1.365(1) 1.361(5) 1.337(6)
B(2)–O(2); 1 1.381(4) 1.377(0) 1.397(7) 1.431(7) 1.363(5)
B(2)–O(3); 2 1.363(3) 1.363(2) 1.370(9) 1.347(7) 1.421(7)
E. Sváb et al. / Optical Materials 34 (2012) 1473–1476 1475
because the neutron scattering amplitudes both for Er (b= 7.79 fm) and for Yb (b= 12.433 fm) are much greater than that for Al (b= 3.449 fm), leading to a significant contrast in the scattered intensities. As a result of the Rietveld calculations we have re- vealed a reasonable good fit for model-1, while model-2 could be excluded. Consequently, we can state that Er3+and Yb3+ions occu- py the Y3+sites similarly to other rare earth elements, and our neu- tron diffraction findings are in good agreement with previous theoretical analysis of the crystal field spectra for Er and Dy doped YAB, which were based on optical spectroscopic experiments[16].
In the Rietveld refinement it was supposed that the dopant Er and Yb elements occupy randomly the Y (3a) sites, as far as no ex- tra Bragg-reflections have been observed as an indication for for- mation of any ordering effect. The occupancy factors were revealed from the nominal compositions, and the actual values were fixed, no refinement was applied. However, it should be noted, that we have tested the effect of refinement of the trivalent lanthanides that share the 3a site with yttrium. We have found that the changes in the occupancy factors did not affect signifi- cantly the atomic position parameters. Concerning the thermal displacement parameters, we have used isotropic displacement, as far as the quality of the experimental data didn’t allow the correct determination of anisotropic thermal parameters of the lanthanides.
The calculations resulted in a fairly good fit for all investigated specimens, as it is illustrated inFig. 1. The positional parameters have been revealed, as they are tabulated inTables 2 and 3. The interatomic distances are important parameters from the point of optical properties. Several atomic neighbour distances are tabu- lated inTable 4.
Our previous spectroscopic studies on the Er-doped YAB crys- tals[6]have shown that Er ions uniformly occupy one specific lat- tice site, trigonal prismatic yttrium position (3a) in accordance with the present neutron diffraction results. Electron paramagnetic resonance studies have confirmed this assumption both for Er and Yb-activated YAB[17]. From luminescence measurements it was established that the decay constant at 10 K was longer (18
l
s) in the Y0.88Er0.12Al3(BO3)4sample than that in the Y0.5Er0.5Al3(BO3)4(16
l
s), which was attributed to Er-Er interactions in the higher Er-content crystal[11].4. Conclusions
Neutron diffraction structure study has been performed on Er and Yb doped YAB compounds and also on a co-doped composi- tion. From the Rietveld refinement of the experimental data the following conclusions can be done:
Er or Yb doped YAB compounds are isotructural to YAl3(BO3)4, the powder pattern can be fitted in s.g.R32 using the triple hex- agonal Wyckoff notation.
Both Er3+and Yb3+ions occupy the Y-sites, and not the Al-sites.
The lattice parameters show a slight decrease with increasing substitution of Er3+and/or Yb3+.
Slight changes are revealed in the positional parameters and in the interatomic distances with increasing concentration of the dopant ions. For the co-doped Y0.84Er0.01Yb0.15Al3(BO3)4 the changes are more significant than for the single-doped YAB compounds.
Acknowledgement
This research project has been supported by the EC-FP7 Grant Agreement N226507-NMI3.
References
[1] A. Brenier, G. Boulon, D. Jaque, J. Garcia-Sole, Opt. Commun. 77 (1990) 221.
[2] V.I. Chani, M.I. Timoshechkin, K. Inoue, K. Shimamura, T. Fukuda, Inorg. Mater.
30 (1994) 1466.
[3] D. Jaque, M.O. Ramirez, I.E. Bausá, J. Garcia Solé, E. Cavalli, A. Speghini, M.
Bettinell, Phys. Rev. B 68 (2003) 5118.
[4] S. García-Revilla, I. Iparraguirre, C. Cascales, J. Azkargorta, R. Balda, M.A.
Illarramendi, M. Al-Saleh, J. Fernández, Opt. Mater. 34 (2011) 461.
[5] N.A. Tolstik, V.E. Kisel, V.N. Kuleshov, K.V. Maltsev, N.I. Leonyuk, Appl. Phys. B 97 (2009) 357.
[6] I. Földvári, E. Beregi, R. Capelletti, A. Baraldi, Phys. Stat. Sol. (C) 2 (2005) 260.
[7] Gy. Mészáros, E. Sváb, E. Beregi, A. Watterich, M. Tóth, Physica B 276–278 (2000) 310.
[8] E.L. Belokoneva, A.V. Azizov, N.I. Leonyuk, M.A. Simonov, N.V. Belov, Zhurnal Structurnoi Khimii 22 (1981) 196.
[9] E. Beregi, E. Hartmann, L. Malicskó, J. Madarász, Cryst. Res. Technol. 34 (1999) 641.
[10] A.A. Kaminski, Lazernie Kristalli Speltroskopicheskie kharakteristiki aktivirovannih kristallov, Nauka, Moskva, 1975.
[11] I. Földvári, E. Beregi, P. Solarz, G. Dominiak-Dzik, W. Ryba-Romanowski, Phys.
Stat. Sol. C 4 (2007) 893.
[12] E. Sváb, F. Deák, Gy. Mészáros, Mater. Sci. Forum. 228–231 (1996) 247.<http://
www.bnc.hu/>.
[13] H. Rauch, L. Koester, Neutron News 3 (3) (1992) 29. <http://www.ati.ac.at/
~neutropt/scattering/Scattering_lengths_table_20010419.pdf>.
[14] M. Fábián, Th. Proffen, U. Ruett, E. Veress, E. Sváb, J. Phys.: Condens. Matter 22 (2010) 404206.
[15] J. Rodriguez-Carvajal, FULLPROF suite <http://www-llb.cea.fr/winplotr/
winplotr.htm>.
[16] A. Baraldi, R. Capelletti, N. Magnani, M. Mazzera, E. Beregi, I. Foldvari, J. Phys C:
Condens. Matter 17 (2005) 6245.
[17] A. Watterich, P. Aleshkevych, T.M. Borowiec, T. Zayarnuk, H. Szymczak, E.
Beregi, L. Kovács, J. Phys. Condens. Matter. 15 (2003) 3323.
1476 E. Sváb et al. / Optical Materials 34 (2012) 1473–1476
